Balancing outputs in a 2-way system

[paradesign]

So, I have tried to do my research and put together a wide range speaker (lows in the 20's) using a two way system, but now that I have everything put together it sounds like crap!

Woofer : Aurum Cantus AC-250F1 10"
Tweeter : Selenium D220Ti-8 with 10" horn

The enclosure is a ported box, with enough volume to have -3db at 29hz.

The woofer is supposed to have a functional range up to 3,000hz, and the tweeter with horn is supposed to have a usable range down to 1,600hz. I am using a P.E. crossover at 2,000hz.

Bing, bang, boom, everything assembles without issue, I wire them to my amp (Marantz 1060b) and the highs are shrill and the bass is nearly non-existent! Hmm, but my amp has some EQ control, so down goes the treble and up goes the bass and things get a little better, hitting the loudness switch turns on the very low bass but things are still miserable.

So, looking back at the driver specs, the Aurum Cantus has a sensitivity of 90db and the Selenium 109db. Is my issue just a problem of mis-matching sensitivities? Or, is there a fundamental problem with building a two-way system of a 10" driver and horn tweeter? Is the Aurum Cantus just not good at driving into the midrange? Can I lower the output of the Selenium to match the Aurum Cantus? Should I have used a lower compression horn? Should I replace the horn with a midrange and tweeter combo to build out a 3-way system?

I plan to do some formal measurements in the next few weeks, but I was hoping to be measuring how close to good I am not how far away...

I'm feeling a bit in over my head, any advice would be appreciated.

[cbdb]

Your horn is almost 20 db more effecient than your woofer (and is that spec in the same box as yours? ) thats a huge difference. Horns are very efficient small woofers in small boxes are not. You should pad the horn down 15 to 20 db (add a resistor in the crossover) and see what that sounds like.

[PeteMcK]

L pad calculator - attenuation dB damping impedance decibel loudspeaker - sengpielaudio Sengpiel Berlin

the xover will also need to deal with the peak in the woofers freq response at ~2.5Khz;

re:'enough volume to have -3db at 29hz' - just how much is this? this driver has a large Vas
(Unibox recommends 150L @28Hz)

[paradesign]

Thank you for the attenuation advice, I will give that a shot first.

Per the peak at 2,500hz in the woofer, will that be an issue if I am crossing over at 2,000hz?

I used the Linear Team tool with the following parameters to design my enclosure:

Vas - 158 / Qts - .33 / Fs - 21

Optimum settings are : Vb - 81.42 / Fb - 25.79
My size constrained (by shelf size) settings were : Vb - 64 / Fb 28

I gave you the wrong number before, the ideal falloff was ~29 @ -3db, and my implementation would hit ~34 @ -3db.

(I acquired Unibox but could not get it to function in MSWord on my Mac, advice on that would also be appreciated...)

Thank you for your assistance.

[PeteMcK]

re:'the peak at 2,500hz in the woofer, will that be an issue if I am crossing over at 2,000hz?' - yes, no matter how steep a crossover you're using it's too close to the crossover point. What you need is a notch filter: Parallel Notch Filter Designer / Calculator Help
Parallel Notch Fliter Designer / Calculator

re:'Vas - 158 ' - where did you get this number from? It's quite different from the published parameter (224L)...

Unibox needs Excel, not Word... it's a lot better than Winisd.

[PeteMcK]

Oh dear, I see you're using an off the shelf crossover... these things are anathema 'round here, for good reason - there's no way they can match the parameters of your drivers, except by pure chance.....

Passive Crossover Network Design

[paradesign]

158L vas came from using the numbers listed at Parts Express... not the datasheet, ugg...

Sorry, I meant Excel, everything Office is MSWord to me...

I receive the following error when I load the "UniBox408.xls" file.

This workbook contains defined names that conflict with valid cell references.

To correct this issue, underscores (_) have been added to the defined names that conflict with cell references. Conflicting names in XLM macro code have not been changed, and must be corrected manually.

I receive the following error when I attempt to do anything.

"The macro '{macronamehere}' cannot be found"

The only utilities I've been able to make work (erroneous data as a separate issue...) are :

Audio Speaker Design Calculators - KBapps.com
and
Linear Team


Apologies for the amateur hour, I am going to head back to the drawing board for a bit before making another attempt. Thank you for the helpful links, I will have my stuff more together when I return.



Before I slink off with my tail between my legs, do you have any recommendations on 10" drivers that work well in enclosures of ~64L? ...that might also be a good match for a highly efficient horn?

[PeteMcK]

Hi paradesign, you may need a newer version of Excel, I've had similar probs in the past with old versions.

Re: recommendations, choosing a more efficient driver is probably a good way to go, but these are usually pro audio and require huge cabs :-(. Someone may be able to recommend a suitable driver

The AC driver in 64L tuned to 30-32Hz will give an F3 of ~44Hz, so all is not lost

[PeteMcK]

an afterthought, take a look at Earls designs here:
Nathan

Although the approach is different, (Sealed Mid-woofer + subs), the B&C drivers may be suitable...???

[Inductor]

Hi.
Pete and others, I don't want to interrupt you guys.

Originally Posted by paradesign

...I am using a P.E. crossover at 2,000hz.

1. You might trash it, unless... a)The xover was designed for a CD, b)You want to save the portion of the woofer xover you have, if you can (because of xover frequencies, bsc and "deal with the peak in the woofers freq response").

2. So I would give a go and start here:
Need good drivers for 'copper' speaker cabinet

3. Would give a look at this xover, compare with the one I (you) have ...and start adapting.

4. And after this, I would have a new design, of course, ...that after a few frequency, phase simulations, would be ready for the test.
Sorry guys, I hate to make work seem so easy.

Note: What type of horn you are using (for reference)?